Polarized distribution of membrane proteins is a characteristic of most differentiated cells and is especially expressed in retinal photoreceptors. We seek to determine to cellular mechanisms of sorting and vectorial distribution of membrane proteins from sites of synthesis to sites of function. Prior studies have concentrated on opsin biosynthesis and demonstrated the role of the Golgi and post-Golgi vesicles in opsin transport to the apex of the rod inner segment. These studies have employed high resolution techniques of immunocytochemistry at the EM level. Antigens are localized in the interior of cells as well as on cell surfaces using polyclonal or mouse monoclonal antibodies on thin sections of retinas embedded in Lowicryl K4M, albumin or frozen sucrose. The distribution of cell surface antigens is being explored by new techniques of ultrahigh resolution scanning electron microscopy using thin (2nm) niobium or tantalum coatings. Bound antibodies are detected by the streptavidin-biotin-bridge technique using biotinyl antibodies and streptavidin-gold or using streptavidin followed by biotinyl-albumin-gold or biotinyl-ferritin conjugates. Labeling densities are quantitated by semiautomated mophometry. Antiopsin antibodies will be used to study rod outer segment disk morphogenesis and terminal stages of opsin transport from the periciliary ridge complex to the outer segment along the connecting cilium plasma membrane. Antibodies to toad retina and brain (Na+, K+) ATPase will be generated and compared for specificity and cross reactivity with presently available polyclonal and monoclonal antibodies against toad kidney ATPase. New monoclonal antibodies will be generated to other photoreceptor and retinal neuron plasma membranes as probes of the sorting of membrane proteins in the Golgi apparatus. Specificity of the antibodies will be further evaluated by binding to electrophoretic transblots of retinal proteins and purified ATPase and its proteolytic fragments. The use of several types of antibodies to several different antigens in photoreceptors and other retinal neurons should allow us to determine the degree to which our findings regarding specific sorting of opsin can be generalized to other retinal membrane proteins. Double label techniques should directly evaluate the role of the Golgi in sorting of these proteins.